Hide table of contents

Purpose of this post:

I hope to share my views on and suggestions for how the strategic decision-making of EA community building (CB) could be improved, and get feedback from the community. More than anything I wish to expand my knowledge about meta EA strategy, build connections, and engage with the community. I am open to feedback, especially from those who think I might be concentrating on the wrong aspects, or who offer insights not directly related to this specific issue but pertinent to community building as a whole. Such perspectives would broaden my understanding of this and other related matters.


This piece might not be perfect and could have several incomplete claims or ideas. I'm open to expanding on them if there's interest. Please feel free to highlight any issues or unclear points. I decided to share this now because I have surpassed my self-imposed deadline and I believe in the value of receiving feedback on imperfect work.

Executive Summary

Based on my experiences with EA community building and mentoring through the CEA Organisational Support Programme, this piece aims to share practical ways to improve community building in the EA movement. It pinpoints the issue of inconsistent understanding and practices within the community and suggests two solutions: launching a Fellowship for community builders and starting a Podcast to share insights. These ideas are meant to unify our knowledge base, welcome diverse thoughts, and enhance the support system for community builders. The need for these initiatives is backed by my work with university groups and discussions with fellow mentors, highlighting the collective push for evidence-backed, effective community engagement.

Caveat: While I believe there are ways to enhance the epistemic health of EA groups overall, this post primarily addresses marginal groups, especially university groups. These groups often struggle with weaker epistemic norms due to frequent leadership changes, alongside new groups that are excited about EA's ideas but lack a deep understanding of the arguments supporting those ideas.

Premise 1: The success of meta EA CB currently relies on many group organisers being naturally capable CBers, correctly understanding the meta strategy and implementing the Operational best practice without having received any particular training or knowledge on how to perform their role, leading to widely varying experiences and epistemic strength among members of different EA groups.

  • Meta: The guiding principles/strategy of what CB hopes to achieve
  • Operational: Best practice and tactical implementation on how to achieve meta-strategies

Premise 2: Community Building involves teachable skills, or best practices, which can be utilised to help many organisers address predictable issues across different cultures and countries.

Premise 3: Providing a framework for scalable opportunities can enhance meta Community Building (CB).

Solutions: Two potential proposals I have thought of which have some historical precedence in the EA space are:

  1. Community-Building Fellowship
  2. Community-Building focused Podcast akin to 80K Hours’


Where I am coming from

Whilst I recognise that my experience in Community Building is not exhaustive, I have been a part of several different communities, ranging from Effective Altruism to sports teams as a player and coach, union branches, and working as a high school teacher. Throughout all of these different groups, I have felt various levels of a culture of ‘Belonging’ similar to what Owen Eastwood speaks to in his book of the same name. Culture is defined by Eastwood as “the environment and the outcome of how we feel when interacting with a group or setting”. Belonging is subsequently a feeling that is then generated by the culture of the group that we interact with. I am glad to say that the closest natural sense of belonging that I have had in a community has been in the Effective Altruism community. 

This came from two places. Firstly, my brother was already in EA and had invited me along over a significant period. However, mostly, it was because, like many who are exposed to the group, I was met with intellectual curiosity and encouraged to explore weird and interesting ideas about how to help others. This exercise, more than anything, is why I continue to be involved with Effective Altruism; I feel that many of the values EA engenders reflect my values. 

However, I have observed, through my readings on the EA Forum, a variety of beliefs and interpretations regarding these values and their implementation within the community. This diversity of understanding appears to be more pronounced on the peripheries of EA and in the realm of Community Building. In these areas, individuals with a limited grasp of the subject matter often engage in recruitment efforts, a process that can resemble the Dunning-Kruger Effect. In such instances, community builders with less epistemic humility may inadvertently compromise the epistemic integrity of EA. Furthermore, these individuals are less likely to receive direct guidance from organisations like CEA. This is partly due to the limited overall benefit their groups might offer and the substantial resources required under the current 1-1 mentorship system to adequately train these organisers before actively supporting their outreach initiatives.

Problems with Inconsistent Community Norms and Epistemics

Effective Altruism (EA) advocates for the Scout Mindset, a value of open-mindedness and a commitment to truth-seeking. However, its practical application within many EA groups appears to be a subject of concern. There seems to be a gap between promoting the Scout Mindset and possessing the skills or comfort level to apply it effectively, particularly when it comes to acknowledging and expressing uncertainties. This issue is highlighted by numerous posts on the EA Forum, where individuals have expressed concerns about the epistemic strength of their local EA community or the EA community at large. The significant level of engagement these posts have received underscores the relevance and urgency of these claims, suggesting a need for deeper introspection and action within the EA community regarding the implementation of the Scout Mindset.

One former EA group organiser from EA Columbia describes EA as ostensibly open but, in practice, seemingly operating under predetermined conclusions. The author suggests that dissent, in coming to a different conclusion, seems to be dismissed as a lack of understanding of the evidence or applying flawed methodologies rather than a valid perspective. This attitude highlights the difference between the external alignment of Scout Mindset values that EA espouses vs the lack of internal alignment of individual organisers or groups. This can lead to weak epistemics and groupthink dynamics, where agreement with certain positions is driven more by social conformity within the group's hierarchy than by a deep understanding of the arguments. With a Karma score of 362 and over 100 comments at the time of writing, many in the EA community seem to resonate with Dave’s concerns. Many of these comments highlight shared similar concerns or experiences, although it is good to note that several comments contest personally sharing the experiences discussed. 

Another post explores how some have become disillusioned with EA. It agrees with the argument that there is incongruence in the internal vs external alignment of Scout Mindset exists, but highlights that those with thoughtful critiques are more likely to leave the community as these critiques go unaddressed, either through lack of pursuit of them and remain silent, or through growing demand for higher quality answers. The author focuses primarily on core EA beliefs and claims around what works best being far less reliable/concrete than what their experience, and those of others they talk to, have witnessed. The author implicitly asks community builders, as community builders are the ones most likely to engage newer community members, to improve their reasoning transparency to minimise the halo effect that some may experience when initially getting involved with EA. While I haven’t noticed this as an issue in my personal experience, it seems as though their concerns again resonate with many in the community, receiving 412 Karma with over 60 comments.

Primary Contributing Factors [CB Practices] 

I think inconsistency in the strength of EA norms and epistemics across different groups can be explained by insufficient support in two places: 1) a lack of specificity about what EA CB is aiming to achieve and how this can be tracked, and 2) lack of practical support for group organisers to use available guidance and resources. 

A lack of specificity about what EA CB is aiming to achieve and how this can be tracked

Largely fueled by a combination of group organisers lacking CB that’s driven to recruit new members as its highest (one of its highest?) priority. The number of Highly Engaged EAs (HEAs) a group can produce is a primary success metric used in much of the rhetoric and advice given by CEA for effective CB, HEA is a placeholder signifying in-depth engagement or personal transformation. The Centre for Effective Altruism (CEA) suggests that the success of EA groups should be measured by the number of Highly Engaged EAs, or “Deep Engagement” on the CEA Community Models page. CEA makes a great effort at this in their Community Models Page in the Groups Resource Center, however, I feel like there is an undervaluation of gathering qualitative data on the internal change process and understanding engaged members as continued engagement does not, in and of itself, signal epistemic strength. University groups especially are going through constant change as people graduate and/or find EA in their second or third year. As a result, there's a tendency to rely on more easily measurable metrics like sign-ups or event attendance. This reliance, as cautioned by Goodhart’s law, risks making these superficial metrics the primary goal. This is not to say that this issue is not known by CEA, but rather it continues as a problem without a clear solution. One other acronym that I have been told better aligns with modern directions for thinking about deep engagement with EA is TATSIS (Talented, Altruistic, Truth-seeking Individuals taking Significant action) but this again is relatively ambiguous.

The EA Forum post 'Community Builders Spend Too Much Time Community Building' highlights how this outreach-centric approach can limit the most engaged members, the Community Builders, from developing skills pertinent to other EA projects or cause areas. This is an even larger counterfactual loss of impact if you consider the 80/20 rule, that 80% of the work is done by 20% of the people, with the Community Builders being the most engaged and likely to fit into the 20% doing 80% of the work/impact of their group. But, in my view, the larger cost of this approach lies not only in neglecting the advancement of knowledge in areas such as AI but also in the missed opportunities for developing more nuanced community-building skills, such as reading and testing out suggestions from Michael Notel’s Forum Post “We all teach, here is how to do it better” or Severin Seehrich’s Lesswrong post “Community Building: Lessons from ten years of facilitation experience”. Focusing on quantitative metrics can eclipse the qualitative growth of the community, hindering the development of skills essential for fostering high-quality engagement and deeper understanding within the EA movement. This was also tackled in a recent post Things newer (university) groups organisers should know about in the section ‘On developing group members’. 

This discussion on whether CB should be Cause or Member focused highlights a critical dichotomy within the EA community. By prioritizing high epistemic standards in member-focused CB, we empower members to make informed decisions about cause areas independently. This approach respects the intellectual capabilities of community members, allowing them to navigate the landscape of causes based on sound reasoning and personal conviction. In contrast, the practice of cause-focused CB risks undermining the epistemic health of our community in the long run. While it might accelerate alignment with certain cause areas deemed important by some, it does so at the potential cost of intellectual diversity and critical thinking.

However, I'm not entirely convinced that those engaging in 'cause first' community building are merely pushing their agendas without offering a balanced perspective. It's not that these community builders lack reasoning or fail to believe in the importance of their causes. Rather, my concern lies with the possibility that they may not sufficiently foster the development of critical reasoning among those they aim to engage. This approach can inadvertently prioritize recruitment over nurturing a deep, reasoned commitment to the cause.

In my mind, EA Community Building and Cause Area field building serve fundamentally different purposes. EA CB focuses on imparting knowledge and autonomy, enabling individuals to determine their best contributions to the world. Here, maintaining epistemic health is paramount. On the other hand, Cause-specific field building is more about expanding the capacity to work within specific areas—recruitment and support for career opportunities take precedence. While both approaches have their merits, the latter's emphasis on recruitment risks sidelining the crucial aspect of why individuals choose to work in those areas, relying instead on aligning them with the philosophical beliefs or findings of the movement.

This distinction raises concerns about cause-specific field building's primary focus. If it leans too heavily on recruitment without fostering an understanding of the underlying reasons for working within a cause, we risk creating a community that operates on alignment rather than informed choice. Such a scenario could detract from the foundational principles of EA, where the ability to critically assess and choose causes based on reasoned judgment is key. In essence, while motivating people to 'just do something' is valuable, it must not come at the expense of the core belief that underpins EA: empowering individuals with the knowledge and reasoning to make informed decisions about their contributions to making the world a better place.

This issue appears structural due to the unclear stance of meta EA CB. The implicit emphasis on members over causes raises transparency concerns. While acknowledging ambiguity, my worry stems from the central aspect of Community Building lacking openness. A comment on Cause or Member-focused CB hints at a "member-first" approach within CEA's work, highlighting the need for clarity and transparency in Effective Altruism Community Building. Without clear communication on the wide variety of different stances that EA CB can take there is a lot of room for misalignment or misimplementation of the nuanced discussions that take place inside of EA and Community Building as a whole. By taking a more deliberate approach it, intuitively, feels like there is a greater opportunity to connect people together. One potential post that explores these thoughts more explicitly is “Rowing and Steering the Effective Altruism Community” which, similarly to the Cause or Member-focused post, highlights the distinct CB efforts on a strategic vs tactical level. Where the strategic (steering) focus is on what the purpose of EA is, and the tactical (rowing) focus is on how to generate growth in power and influence. Using some of the language from the post, I would suggest that I highlight the Community's failure to steer. “Community failures concerning culture, (epistemic) norms, homogeneity of worldviews or demographics, whether the movement should be large or narrow, etc.” Lastly, the post, Bad Omens in Current Community Building, which self-identifies as attempting to provide some ‘steering’ on the issue of recruitment at the expense of epistemic strength continues to suggest this is an issue felt strongly by the EA community. The post received a Karma score of 568 on 346 votes and over 130 comments (at the time of writing).

Lack of practical support for group organisers to use available guidance and resources. 

What resources currently exist?

  • CEA run 1-1 Mentoring/Coaching
    • Organisational Support Programme
    • University Group Accelerator Programme
  • Freely Accessible Resources
    • EA Forum
    • Groups Resource Hub
    • Groups Slack
    • EA Groups Newsletter
  • Tangentially related to community-building skills/support
    • Effective Thesis
    • 80K Hours website + Podcast

I'd like to extend my appreciation for the substantial efforts of CEA in deploying resources effectively to support university groups around the globe. The role of mentors, with their expertise in Community Building, is undeniably pivotal in this process. It’s understandable that, given the large number of potential mentees, mentors might face certain challenges. This situation, while natural in large-scale mentorship programs, could lead to variations in the mentorship experience.

It's also worth noting that the process for standardising the approach among mentors doesn’t appear to be very stringent. This observation is not a criticism but an acknowledgment of the complexity and diversity of the mentorship landscape. Such flexibility can have advantages, allowing for a more personalised mentorship approach. However, it also suggests an opportunity for further refinement, perhaps by exploring balanced standardisation methods that respect the individuality of mentor-mentee relationships while ensuring a consistent level of guidance and support.

In light of this, along with anecdotal reports that some mentors are engaging with a significant number of mentees – at times up to 15, accounting for nearly half of their working hours – it becomes important to consider the efficiency and effectiveness of the current system. While the dedication shown by these mentors is commendable, there may be potential to streamline the process in a way that maximises the impact of their time and effort. Such an approach would not only support the mentors in their roles but also enhance the overall quality and impact of the mentorship provided.


Owen Eastwood, the author of Belonging, emphasises the importance of connecting people ritually and establishing standards for the vision and values of Community Building. While Eastwood primarily operates in diverse fields, notably within sports teams, he underscores that a successful group or organisation hinges on a sense of Belonging, clear communication and personalisation of values, things that are reflected in Michael’s post linked earlier. In my view, Effective Altruism excels in articulating and embodying these values in interactions with highly engaged individuals, such as at a retreat or conference. However, this success has not been fully extended to broader outreach, where most Community Builders operate. The current approach offers limited options, such as occasional meetings for those not near a core centre or, in worst cases, leaves individuals stumbling in the dark. Resources are heavily concentrated on promoting 1-1 (in-person or virtual) interactions supporting top colleges/cities globally. While recognizing the merit in targeting high academic achievers, it remains unclear whether this focus on a select demographic outweighs the potential loss of impact that a more nuanced and inclusive development approach could bring.

EA CB Fellowship:

A potential solution to enhance community building (CB) within Effective Altruism (EA) could be the establishment of a fellowship program specifically for CB. This program would aim to define essential skills, present arguments for and against various approaches, and offer career guidance for those interested in this sector. Given the existence of fellowships for other EA cause areas, this initiative could elevate CB to a similar level of importance and recognition. It’s a timely proposition, as there's a strong argument for emphasising CB at this stage. My understanding is that there was an attempt to create something akin to this with the pilot running this syllabus. I have been told that it was initially scrapped as a project as it was not getting the results that it desired, however, I have no information on the details of the project, its metrics for success, or the level of achievement. Interestingly, many of the posts I have linked can be found in the syllabus.

EA CB Focused Podcast:

Another solution might be to launch a podcast similar to the 80K Podcast, focusing on CB. Podcasts are increasingly popular and accessible, fitting easily into people’s daily routines. This medium could disseminate advice and insights from CB leaders, standardising knowledge and skills in the field while avoiding the creation of a uniform CB culture. This approach could enhance the quality of engagement in the community and address recent concerns about community health. Examples of how these have been useful include the 80K Hours episode with Benjamin Todd, Benjamin Todd on what the effective altruism community most needs (& how to analyse replaceability) and the After Hours podcast episode with Kuhan Jeyapragasan on effective altruism university groups. There are many more episodes that would be helpful for Community Builders, but don’t necessarily deal with the EA Community or the act of Community Building as a whole.

Having served as a mentor for the CEA program, I’ve observed a tendency towards a one-size-fits-all approach in CB, with minor adaptations for different groups worldwide. I wish to make it clear that I do not believe that there is any issue with this approach, on the contrary, it is quite reassuring that the issues that group organisers, usually university, have a predictable framework of issues from which to build a comprehensive response. A more deliberately crafted framework allows for consistency yet encourages adaptation and iteration. This could provide a clearer understanding of each group's impact and facilitate improvements over time. By adopting a general rule in the face of uncertainty and continuously refining our approach as new information becomes available, we can increase the chance of a more effective and responsive strategy in community building.


I'm aiming to significantly enhance the pool of high-quality community builders within the EA community, tackling the critical shortage of resources and mentors available to support our growing number of community builders. Despite the trend that those organizers most in need often receive the least support, my strategy is designed to broaden the support network through our fellowship program, thus elevating the foundational level of support while still maintaining the upper echelons of excellence. This approach is based on a theory of change acknowledging the existing demand for community-building support and the demonstrated benefits of such support, as evidenced by initiatives like the OSP by CEA, 80K's coaching, and various virtual programs and podcasts. My objective is to enhance the impact of both top-performing and smaller groups while mitigating the risks associated with potential mismanagement. While further evidence is necessary to wholly validate this approach, the potential benefits of optimizing mentor time and providing scalable resources make this initiative worth pursuing in my view.

It's crucial for the Effective Altruism (EA) community to thoughtfully consider and engage with the concerns that may arise within its broader groups. Addressing these matters attentively can help us avoid potential challenges, such as unintentionally impacting the community's reputation, which in turn could influence engagement levels and the rich diversity of contributions we value so highly. Emphasizing the importance of maintaining consistent community norms and mentorship processes will bolster our outreach efforts, ensuring we continue to attract a wide array of perspectives. This diversity enriches our community, upholding the high epistemic standards central to EA's philosophy. By embracing strategic, inclusive, and well-structured community-building efforts, we can support EA's growth and its ongoing mission to drive meaningful change. This approach not only preserves our core values but also strengthens our collective capacity to make a significant impact.


One of the key uncertainties in my understanding of Effective Altruism’s (EA) community-building practices is the actual impact of prioritising recruitment and numerical metrics over deeper engagement. To address this uncertainty, there is a crucial need for more empirical data from community engagement studies. Such data could provide invaluable insights into how this approach affects the community's depth of engagement, personal transformation, and overall health. My lack of empirical evidence presents a gap in my knowledge, making it challenging to assess the effectiveness of current strategies and to make informed decisions about potential improvements.

When proposing solutions like a fellowship program for community building and a podcast akin to the 80K Podcast, I base these on the observed success of similar initiatives in other fields. For instance, I base the continued implementation of other EA fellowships and virtual programmes on the belief that they have a significant impact in helping participants pursue impact and careers in their respective fields, but I don’t have specific data on how successful these efforts have been. It's crucial to also consider counterarguments, such as the risk of oversaturating the EA space with too many specialised programs or the potential for podcasts to disseminate a uniform culture, which could stifle diversity in thought and approach.

Additionally, in advocating for a more nuanced framework in community building, I draw from my experiences with the Organisational Support Programme. While these experiences are valuable, I acknowledge the need to juxtapose them with a wider range of perspectives from the EA community. It's important to consider whether these issues are universally recognized within the community or if they might be interpreted differently by others. This is why I turn to you, the EA Community, to begin this journey and help fill the gaps in my knowledge and experience.

I have also made claims about various EA Forum posts and using metrics such as Karma and comment engagment to highlight the level of concern the community has about this issue. It is worth noting that these posts could be representative of the same voices who are consistently sharing concern of these issues rather than every vote representing a different person. It is also difficutl to tell whether the support for these posts is representative of this being an existing issue, or merely reflects the communities sensitivity to not wanting this to become an issue by highlighting it as a concern.


The discussion suggests an assumption that the issues raised are indeed significant problems within the fringe Effective Altruism community's epistemic status. However, this assumption may require further scrutiny and validation. The evidence presented so far primarily stems from the personal experiences and posts of individual group organisers, and despite the posts engagement on the EA Forum it may not be enough to generalise about the value alignment and epistemic quality of EA community building as a whole. This point is critical to my argument; if these issues aren't as widespread as assumed, the proposed solutions might not be addressing the core problems, or it may not be a problem at all and instead a shared concern for this being a potential issue. To gain a clearer understanding and personally apply the Scout Mindset in these assertions, further research is needed, including:

  1. Gathering personal experiences as a group organiser, both locally and within the broader EA community, such as at EAG(x) events.
  2. Examining the discussions and feedback from others who have either organised groups or participated in university clubs, particularly on platforms like the EA Forum (comment below)
  3. Consulting with immediate contacts, such as OSP mentors and the CEA Groups Team, to gather their perspectives and experiences.
  4. Consider conducting a comprehensive survey of community members involved in university groups, including members, organisers, and other stakeholders like funders. This research could help validate or challenge my current assumptions and should be approached with a Scout Mindset, ready to update beliefs based on the strength and relevance of the evidence gathered. This survey could also be a valuable first step in assessing the generalizability of the issues I’ve raised and help determine whether the proposed solutions are appropriately targeted. It’s important to consider that even if the evidence suggests that these issues are more isolated or specific to groups with certain characteristics, it doesn't necessarily render the proposed solutions irrelevant but may require a more tailored approach.

In conclusion, while my suggestions aim to foster a more effective and responsive strategy in EA community building, incorporating a diverse range of experiences and data, along with a consideration of different viewpoints, will enrich our understanding and approach to these complex issues.





More posts like this

Sorted by Click to highlight new comments since: Today at 4:39 AM

I would love to see attempts at either a community-building fellowship or a community-building podcast.

With the community-building podcast, I suspect that people would prefer something that covers topics relatively quickly as community builders are already pretty busy.

Curated and popular this week
Relevant opportunities